New injecting drug users (IDUs) are at risk of becoming infected with the hepatitis C virus (HCV), the hepatitis B virus (HBV) and HIV within the first few years after they start to inject. However, the risk of infection among new injectors varies by race/ethnicity, with African Americans at greater risk of infection than white new IDUs. Risk behaviors alone do not sufficiently explain why African American IDUs are at greater risk of infection. Racial/ethnic differences in the risk of infection among new injectors may be related to different probabilities in the risk of exposure to infection, which may stem from differences in sociocentric (groups of directly and indirectly connected IDUs) and egocentric (groups of directly connected IDUs) transmission networks. Variation by race/ethnicity in the transmission networks of new IDUs may also be influenced by neighborhood differences in the provision of harm reduction and other health services, as well as other differences in the social and physical environment of neighborhoods, since IDUs are often geographically separated by race/ethnicity into different neighborhoods. In this study, we will investigate the sociocentric and egocentric network risks for infection with HIV, HCV and HBV in two high-risk neighborhoods in New York City, one of which is predominantly African-American and the other in which new IDUs are predominantly white. Using a sequential cross-section design with an imbedded prospective cohort, we will compare a predominantly African-American group of new injectors and a predominantly white group of new injectors m geographically separate neighborhoods to determine differences in: (1) The prevalence and incidence of HIV, HCV and HBV; (2) The relationship of sociocentric and egocentric networks to HIV, HCV, and HBV transmission; (3) The relationship of sociocentric and egocentric networks to risk and risk reduction behaviors, knowledge of infection risk and prevention, and access to and utilization of harm reduction and other health services; (4) Factors associated with the maintenance or change in sociocentric and egocentric networks and risk and risk reduction behaviors and practices, and the relationship of these factors to the rate of HIV, HCV, and HBV seroconversions; and (5) using ethnographic methods and Geographic Information Systems, to explore the relationship of neighborhood factors to network and behavioral infection risk. Understanding the extent to which the multi-level influences of neighborhood context, network characteristics and behavior shape the risk of acquiring or transmitting HIV, HCV and HBV among new injectors can contribute towards the development of structural interventions targeted at new injectors and their injecting and sex partners in high-risk neighborhoods. [unreadable] [unreadable]